Exploring the potential of using optical log scanners for predicting lumber grade

Swedish softwood sawmills have traditionally sorted logs into relatively homogeneous size classes to facilitate downstream production processes. Sorting is normally based on top-end diameter classes as determined by optical log scanners. The purpose of this study was to determine the feasibility of using conventional optical log scanners to determine log quality. A total of 300 debarked Scots pine (Pinus sylvestris) sawlogs delivered to a large sawmill in northern Sweden were scanned using the dual-axis scanner installed at the mill. Log profiles, consisting of diameters measured in two directions to the nearest millimeter every second centimeter along the log length, were generated by the scanner and stored on diskette. The scanned logs were sawn and the resulting lumber graded. The scanned profile data were used to develop computer algorithms describing various parameters of log geometry including: taper in different sections of the log, surface roughness, sweep, and eccentricity. Parameters describing the shape of each log were combined with the lumber grade information to determine which parameters indicated lumber quality. Taper in the large end was found to be a good indicator of log position, which can indicate lumber grade. Surface roughness was related to the grade within position classes, but neither sweep nor eccentricity demonstrated any relationship with lumber grade. These results have been used to develop a log sorting algorithm to automatically identify and sort logs with the geometric features associated with grade recovery.

You must be logged in to download any documents. Please login (login accounts are free) or learn how to Become a Member